While controversies continue to develop with regard to the surprising results of the United Kingdom referendum to determine whether or not to stay within the European Union, we are more interested in visualizing and exploring the data surrounding the vote. Our first post on the Brexit showed you the mechanics of how you can connect the ballot results with data on geography, income, and education across the UK. Our second post took a look at the geography to better understand where voters were coming from in this historic decision. This, our third post, will focus on insights into how education levels may have affected whether a person voted to leave or remain in the European Union.
For those of you just joining us, the UK Election Commission website has rich publicly available data on poll results. Our income data, which we will explore in a later post, came from the Office for National Statistics site and we added education data at the region and area level for Scotland, England and Wales, and Ireland, respectively. This was an hour or so of Googling to find the data we needed. We then loaded the data into Arcadia and modeled relationships between key fields in the data to begin visualizing the information to gain insight into voting results.
So let’s tease apart the data by level of education. It turns out level of education of the population in a region correlates strongly with their Brexit vote. The scatter plot visual above clearly shows the impact of education on the desire to remain in or leave the EU. The greater the number of residents with a higher level of education, the more likely that local authority was to vote to remain in the European Union.
This is most evident in central London, where more than two-thirds of the city population has a bachelor’s degree. The East Midlands has the highest percent of the population without a degree (~30%) and the highest percent leave vote. Remember, it could also be that a higher concentration of people with higher education levels is a proxy for something else, urban versus rural for instance.
In addition to London, Scotland is also an outlier. The way the Scottish voted wasn’t correlated with income and education like in the rest of Britain. People there voted to remain in the European Union regardless of their education or income. And there is talk among the Scottish leadership about fighting to remain in the European Union even as Britain leaves.
As with most data visualizing, there is a caveat: correlation doesn’t correspond with causation. This is more hypothesizing than deducing results from the broad geographic and census data we gathered and always carries a chance of error. There’s always other data. For example, we didn’t take religion into account, and the number of Jedi knights in a given locality may have influenced the ‘Remain’ vote…but only in a galaxy far, far away.